空间数据挖掘技术方法及应用

着重阐述了通用的空间数据挖掘体系结构,空间数据的关联特性,几种主要的空间数据挖掘方法。最后对一实例进行了应用分析。

The Technology and Methods of Spatial Data Mining and Their Application

Remotely sensed data are so large that they can not be dealed with by human brains. Therefore, it becomes increasely important to find a method to get information automatically, quickly, efficiently from remotely sensed data. The technology of Data Mining was put up with in 1980's, and it has been used in many fields, such as finance fields, insurance fields, retail fields, medical and justice, etc. Data mining is the non-trivial process of identifying valid, novel, potentially useful, ultimately understandable patterns from huge volume of data. The spatial data mining is developed on the base of it. In a sense, spatial data mining resolves the problem of abundant data that human brain cannot be competent with. Furthermore, it can be used in artificial intelligence and pattern recognition fields. Spatial data mining differentiates from general data mining. Because Spatial data mining must consider the topology relation between the neighboring object, the direction and distance, etc. On the contrary, the general data mining consider every element in the database is independent. Nowadays, although the technology of the spatial data mining develops very quickly, it has not matured. There still are many things for us to do, especially, the extraction of the hedge and the feeble information from the remotely sensed images. This paper provides the background of spatial data mining, the general process of data mining, common algorithms, language and prospect.